Double layer metaheuristic based energy management strategy for a Fuel Cell/Ultra-Capacitor hybrid electric vehicle

This paper highlights the use of metaheuristic approaches to perform the energy management of a hybrid Fuel Cell/Ultra-Capacitor Electric Vehicle considering hydrogen consumption, Fuel Cell durability and computational time as key performance criteria. The considered architecture is an integrated rule-based metaheuristic approach that combines the simplicity and the effectiveness of rule based and optimization approaches. Online results compared with benchmark offline Ant Colony Optimization results, present near to optimal performance, with slightly better performance for Particle Swarm Optimization versus Genetic Algorithm. The durability of the main energy source which is a crucial issue for Fuel Cell Hybrid Electric Vehicles is considered by limiting the Fuel Cell power variation rate in both layers providing a high protection level in order to respect its slow dynamics that represent a critical operational limitation. The rule-based layer which is a strategic layer reduces computational effort and time and restricts the search space of the optimization layer allowing a rapid convergence and thus enabling real time applications. Integrated Particle Swarm Optimization algorithm achieved lower computational time with an average of 0.65 ms versus 43.09 ms for the integrated Genetic Algorithm for each time step, which makes it more suitable for real time applications.